Chapter 5: Demand Forecasting & Collaborative Planning,
Forecasting, & Replenishment
Principles of Supply Chain Management: A Balanced Approach
Prepared by: Daniel A. Glaser-Segura, PhD
Introduction to Forecasting
1. What is Forecasting?
o Forecasting is the process of estimating future demand for a product or service.
o The goal of forecasting is to predict future demand as accurately as possible,
minimizing any forecast errors.
2. Why is Forecasting Important?
o Forecasting helps businesses align supply with demand, ensuring that resources (like
inventory and production) are planned efficiently.
o An accurate forecast allows companies to make decisions that minimize costs, such
as optimizing inventory levels and production schedules.
3. Factors Influencing Demand:
o Market trends (e.g., consumer preferences)
o Economic conditions (e.g., inflation, unemployment)
o Competitor actions (e.g., new product launches, price changes)
o External factors (e.g., weather, political events)
o Understanding how these factors influence demand is crucial for forecasting.
4. Benefits of Accurate Forecasting:
o Lower inventories: Helps reduce storage costs by only keeping the necessary
amount of stock.
o Fewer stock-outs: Minimizes the risk of running out of stock, which can lead to lost
sales and dissatisfied customers.
o Smoother production plans: Allows for consistent and efficient production without
excessive disruptions.
o Reduced costs: By avoiding overproduction or underproduction, businesses can
reduce waste and improve operational efficiency.
o Improved customer service: Ensures that products are available when customers
need them, increasing satisfaction.
Matching Supply and Demand
1. The Role of Suppliers:
o Suppliers need to forecast demand to ensure they can provide the right quantities
at the right time and at the right cost.
o They must balance production cost, product quality, and customer service to stay
competitive in the market.
2. Challenges:
o Any disruption in production or delivery can impact the entire supply chain. For
example, delays or shortages can cause inventory issues downstream.
Forecasting Techniques
There are two broad categories of forecasting methods: Qualitative and Quantitative.
1. Qualitative Forecasting
When to use: Qualitative forecasting is ideal when data is limited, unavailable, or irrelevant.
It relies heavily on the judgment and expertise of people involved in the forecasting process,
such as managers or industry experts.
Methods of Qualitative Forecasting:
1. Jury of Executive Opinion: A group of experts in the field discusses and forecasts
demand based on their experience and judgment.
2. Delphi Method: Similar to the Jury of Executive Opinion, but feedback is gathered
independently and anonymously from experts in several rounds, with the results
aggregated each time.
3. Sales Force Composite: Salespeople use their knowledge of customers to forecast
future demand. They provide individual forecasts, which are then combined into a
collective forecast.
4. Consumer Survey: Direct feedback from consumers is gathered through surveys or
interviews to predict future purchasing behavior.
2. Quantitative Forecasting
When to use: Quantitative forecasting is more appropriate when data is available. It relies
on historical data and statistical models to predict future demand.
Types of Quantitative Forecasting:
1. Time Series Forecasting: Assumes that past demand patterns will continue in the
future. It uses historical data to identify patterns and trends.
2. Associative Forecasting: Involves identifying factors (independent variables) that
influence future demand, such as marketing campaigns, price changes, or economic
indicators.
Recommendation: It is often beneficial to use a combination of qualitative and quantitative
methods for more accurate forecasts.
Time Series Data and Components
Time series forecasting looks at historical data to identify patterns and make predictions. These
patterns can be broken down into four key components:
1. Trend Variations:
o The long-term direction in demand over time. This could be an increase or decrease
in demand.
2. Cyclical Variations:
o These are long-term, wave-like patterns that often repeat over a period longer than
one year. For example, economic cycles or industry-specific trends.
3. Seasonal Variations:
o Regular, short-term fluctuations in demand that happen at predictable intervals.
For instance, higher demand for ice cream during the summer or higher retail sales
during the holiday season.
4. Random Variations:
o These are unpredictable changes in demand caused by unexpected events such as a
natural disaster, strikes, or sudden market shifts.
Forecasting Models Based on Time Series
1. Simple Moving Average:
o This model forecasts demand by averaging past data over a set number of periods
(e.g., the past 3 months).
o It is effective when demand is stable and doesn’t have large fluctuations.
2. Weighted Moving Average:
o This model is similar to the simple moving average but assigns different weights to
different periods, giving more weight to the most recent data.
o This method is useful when recent trends are expected to be more indicative of
future demand than older data.
Forecast Accuracy
To measure how accurate a forecast is, forecast error is calculated:
Forecast Error(et)=At−Ft\text{Forecast Error} (e_t) = A_t - F_tForecast Error(et)=At−Ft
Where:
ete_tet = Forecast error for period ttt
AtA_tAt = Actual demand for period ttt
FtF_tFt = Forecast demand for period ttt
Key Metrics to Measure Forecast Accuracy:
1. Mean Absolute Deviation (MAD):
o MAD measures the average of the absolute differences between the actual demand
and the forecasted demand. A lower MAD indicates a more accurate forecast.
o A MAD of 0 means the forecast was perfect.
2. Mean Absolute Percentage Error (MAPE):
o MAPE calculates the percentage of forecast error. It is the average of the absolute
percentage errors over a period. A lower MAPE indicates a better forecast.
Collaborative Planning, Forecasting, and Replenishment (CPFR)
CPFR is a collaborative process that helps supply chain partners work together to plan key activities,
from the production of raw materials to the final delivery of products to customers. This
collaboration leads to more accurate forecasts and greater supply chain efficiency.
Objectives of CPFR:
Optimize the Supply Chain: Improve forecasting accuracy, minimize inventory, reduce stock-
outs, and ensure products are delivered at the right time and place.
Enhance Customer Service: By improving forecasts, the right products are available when
customers need them, increasing satisfaction.
Value of CPFR:
By sharing information on sales trends, promotions, store openings/closings, and new
product launches, CPFR improves forecasting accuracy.
Collaboration ensures that all parties have aligned expectations, leading to better decision-
making and smoother operations.
Steps in the CPFR Process (Voluntary Interindustry Commerce Standards Process Model)
The VICS CPFR Model provides a structured framework for CPFR implementation, involving several
key steps:
1. Develop Collaboration Arrangement:
o Set up the framework for cooperation between supply chain partners.
2. Create Joint Business Plan:
o Develop a shared business plan outlining goals, objectives, and strategies for
collaboration.
3. Create Sales Forecast:
o Use historical sales data and other inputs to create a forecast of future demand.
4. Identify Exceptions for Sales Forecast:
o Detect discrepancies between the forecast and actual sales, identifying areas for
improvement.
5. Resolve/Collaborate on Exception Items:
o Collaboratively address any exceptions or unexpected events impacting the forecast.
6. Create Order Forecast:
o Based on the sales forecast, develop a forecast for the amount of stock that needs
to be ordered.
7. Identify Exceptions for Order Forecast:
o Check for any issues with the order forecast, such as inventory shortfalls or supply
disruptions.
8. Resolve/Collaborate on Exception Items:
o Collaboratively resolve any discrepancies in the order forecast.
9. Order Generation:
o Finally, generate the orders needed to replenish stock, ensuring that the right
products are available at the right time.
